--On 6. Januar 2007 12:15:19 -0500 Stephan Richter <[EMAIL PROTECTED]> wrote:

On Saturday 06 January 2007 12:03, Andreas Jung wrote:
returns always unicode for non-strings but keeps standard Python
strings as they are.

I think we always require to return unicode text within Zope 3,

which is a good policy

so there
should never be a regular string in there.

this policy must be enforced? :-)

I have some ideas how make resolving UnicodeDecode
error configurable but I am not sure if this code should go into the
Zope 3 or should remain in the Zope 2 core. Does Zope 3 has to deal with
such kind UnicodeDecode errors?

I think this is a pure Zope 2 problem at this point, no? We certainly
have way  of handling UnicodeDecode errors. If you can show that this
could potentially  be a problem in Zope 3 as well, I would be for adding
your UnicodeDecode  resolver to the Zope 3 core.

It's possibly a major Zope 2 issue. However since we are putting more and more Zope 3 stuff into Zope 2 is becomes also (in some way) a Zope 3 issue. I just wonder if it would be better to move more logic into the related Zope 2 wrapper code (making it more fat) or putting it into the Zope 3 core...
I have to brainstorm about it.

BTW, weren't you in the room when we had the initial discussions pretty
exactely 5 years ago?

Yes, you were! :-)


Oh my god :-) Yet some more years we survived with Zope :-)


Attachment: pgp7USG8D73j1.pgp
Description: PGP signature

Zope3-dev mailing list
Unsub: http://mail.zope.org/mailman/options/zope3-dev/archive%40mail-archive.com

Reply via email to